Puberty differentially predicts brain maturation in male and female youth: A longitudinal ABCD Study

Abstract

Research has demonstrated associations between pubertal development and brain maturation. However, existing studies have been limited by small samples, cross-sectional designs, and inconclusive findings regarding directionality of effects and sex differences. We examined the longitudinal temporal coupling of puberty status assessed using the Pubertal Development Scale (PDS) and magnetic resonance imaging (MRI)-based grey and white matter brain structure. Our sample consisted of 8896 children and adolescents at baseline (mean age = 9.9) and 6099 at follow-up (mean age = 11.9) from the Adolescent Brain and Cognitive Development (ABCD) Study cohort. Applying multigroup Bivariate Latent Change Score (BLCS) models, we found that baseline PDS predicted the rate of change in cortical thickness among females and rate of change in cortical surface area for both males and females. We also found a correlation between baseline PDS and surface area and co-occurring changes over time in males. Diffusion tensor imaging (DTI) analyses revealed correlated change between PDS and fractional anisotropy (FA) for both males and females, but no significant associations for mean diffusivity (MD). Our results suggest that pubertal status predicts cortical maturation, and that the strength of the associations differ between sex. Further research spanning the entire duration of puberty is needed to understand the extent and contribution of pubertal development on the youth brain.

Keywords: ABCD Study; Adolescence; Brain maturation; Development; Longitudinal; Puberty.

Fig. 1

Age (left) and PDS (right) distribution of the study sample, split by sex and timepoint. Dotted line between plots connects the mean age and PDS values at baseline (timepoint 1) and follow-up (timepoint 2). Age and total PDS scores shown as distributions where green represents females and purple represents males.

Fig. 2

Bivariate latent change score (BLCS) model with two timepoints, examining correlation between pubertal status (PDS) and brain MRI metric (MRI) at baseline (purple), to what extent pubertal status at baseline predicts the rate of change in brain MRI metric (blue: $\gamma$1), to what extent brain MRI metric at baseline predicts the rate of change in pubertal status (blue:$\gamma$2), and whether changes in PDS and MRI co-occur (yellow) after considering the coupling ($\gamma$1, $\gamma$2) pathways. Image adapted from Kievit et al. (2018) with permission from author.

Fig. 3

Graph showing the multigroup BLCS models for fractional anisotropy (FA) and PDS relations.

Fig. 4

Graph showing the multigroup BLCS models for mean diffusivity (MD) and PDS relations.

Fig. 5

Graph showing the multigroup BLCS models for cortical thickness (CT) and PDS relations.

Fig. 6

Graph showing the multigroup BLCS models for surface area (SA) and PDS relations.